Paramount first studio to end distributing celluloid

[Gunleik Groven]

... and it can be automated.

But does not work without a significant interpreter level...

 

[Dominik Muench]

In terms of quality isn't that great news ?
For digitally acquisitioned films, the 35mm print was always the weakest link at the end regarding projection quality, nothing worse than seeing a film shot on 4K on a dim, dusty, uncalibrated and unsteady film projector.....now that its all digital, without quality loss in the copying process, going to digital cinema packages and full 4K projection we finally have a consistent shooting to projection workflow ^^

 

[dio Zafi]

If you've ever seen first-gen prints straight off a neg projected big and beautiful you cry at how much is just gone by the time it hits the multiplex. This is a big reason why the death of film as an acquisition format is far too premature, regardless of how good any digital camera is, because DCP of analog origination is the best looking digital presentation going. Similarly, digital origination looks nice® when printed to film and optically projected. It would be interesting to see experiments with digital origination, print to neg and then scan that negative for DI and eventual DCP.


this is a quote from a TEXAS filmmaker…

 

[Dominik Muench]

Dio: all nice and well....but 99% of the people will never see a first-gen print that way....so whats the point ?

Similarly, digital origination looks nice® when printed to film and optically projected.

Matter of opinion on which I tend to disagree.

Most people that go to the cinemas...unless they have an immaculately maintained big $$$ multiplex in their neighbourhood will watch movies in a small theatre with averagely maintained projectors where the projection light is dimmed to save money.

A few years ago I shot an interview with the amazing Don McAlpine ACS/ASC where he discusses exactly THAT aspect about digitally shot films and projection quality of 35mm release prints.

Fast forward to 20:10 and listen to what he has to say.

Further down he says that for him...a true full 4K digital workflow from acquisition to projection cant come soon enough.....and this coming from a master who shot big budget 35mm projects his entire life.

 

[Paul Schneider]

Been saying it for years... Now is starting to sink in to the rest... The nostalgia will also not last long...

Not as a capture medium, but with regard to long term archival film is (and probably should remain) the only game in the business.

 

[Gavin Greenwalt]

Today, that's not so difficult because the formats are usually some kind of current standard. In the studios' case, they only use open standards precisely for that reason. But what they're written on has a format (LTO, optical storage, etc) and the files themselves have a format (DPX, EXR, even things like R3D that I wouldn't recommend for archival storage due to their proprietary and undocumented nature). I doubt that you're putting instructions as to how the data was written and how to interpret it on every one of those copies.

I hear this over and over... it's even on Kodak's Advertisements--but again it's total nonsense. ZIP Disks are now 20 years old. It's proprietary and was never intended for long term archival. But it was a mass produced consumer product. As a result--it's trivial to find a working drive. Not only that you can literally buy them by the palette load. I could find a functional Zip Drive and hook it up by the end of the day. I wouldn't say that I could get a full scan of a separation archival film reel. When you produce millions of consumer devices there will almost always be cheap working examples. Need an 8 track player? No problem. It's 40 years old but there is probably one for sale within a mile radius of you right now that's in pristine condition for less than $100.

USB is now nearly 20 years old. And yet you can take any USB device from 1996 and plug it into almost every single computer in existence. FAT the file system has been around since the early 1980s. It's the formatting used on REDMags. Optical storage is an open ISO standard. I mean even the "O" in LTO tape is "Open". If you are truly worried than you can just archive a few sealed boxes in your vault.

Are they all unproven in the long term? Yes. Because they haven't existed in the long term. However, seat belts hadn't been proven in the long term either before we could conclude that they increased safety.

In my opinion it's the stone vs paper battle. The film industry is firmly entrenched in the "Stone" camp. They think it's worth carving it into stone since stone is strong. Stone is proven to last centuries, stone is expensive but good. I'm in the paper camp. Paper is flimsy. Paper burns. Paper is spoileable by water and animals. You would think stone is the way to go--but if you want Harry Potter to survive 1,000 years you don't put it on an elaborate stone tablet system, you print 10,000,000 copies and it'll just happen. Volume trumps strength.

I've also seen countless 20th Anniversary DVDs where they say the same thing every time. "The film was falling apart, if we didn't do it now we were going to lose it forever." I start from the assumption that people are going to screw up. And then based on that assumption what is most resistant to cock-ups? From the NASA archives to classic films the answer is always the exact same: some guy had a copy in his sock drawer or some TV station in Melbourne happened to have an old copy. Cockroaches are forever, not diamonds--diamonds get dropped onto the Titanic Wreckage never to be seen again.

 

[George Butterfield]

I am sometimes playing head-games with what an archaeologist will do when he finds a CD,

Probably stick it in the CD player he found beside it.

 

[Gavin Greenwalt]

Probably stick it in the CD player he found beside it.

It's also a little presumptuous that we need to worry about Archaeologists. Unless we get hit by a comet, die of a zombie plague or somehow or another lose the internet in some apocalypse--we're retaining an extraordinary amount of knowledge. Sure if some future archaeologist finds I Love Lucy on DVD they won't be able to play it--but if a future archaeologist in 1,000 years has lost the ISO standard somehow I'm sure we've also lost any film masters. Archaeologists largely only find our trash. Masterpieces generally don't last very long unless they're marble.

It's the same mistake that Gold fetishists have. "When the Apocalypse strikes your cash will be useless! But I'll have a suitcase full of Gold and be fine!" Completely ignoring the fact that if the apocalypse strikes currency will collapse period and if you can't eat it or wear it it's probably useless. Similarly if a comet strikes and we forget how to read a DVD... will we really be guarding our film vaults and keeping them humidity controlled? If we're scavenging for food on a irradiated apocalyptic wasteland... is it really that important that we don't forget LOST? So much of what we do I'm sure would be interesting to historians but that far removed from our society I doubt Seinfeld will resonate the same with a hunter gatherer culture.

 

[Marc Wielage]

I hear this over and over... it's even on Kodak's Advertisements--but again it's total nonsense. ZIP Disks are now 20 years old. It's proprietary and was never intended for long term archival. But it was a mass produced consumer product. As a result--it's trivial to find a working drive. Not only that you can literally buy them by the palette load. I could find a functional Zip Drive and hook it up by the end of the day. I wouldn't say that I could get a full scan of a separation archival film reel. When you produce millions of consumer devices there will almost always be cheap working examples. Need an 8 track player? No problem. It's 40 years old but there is probably one for sale within a mile radius of you right now that's in pristine condition for less than $100.

The problem you overlook is that the LTO tapes and drives both can fail. And at the moment, this is all we have for longterm digital storage. I have seen multiple copies fail at different times, and it's not pretty when that happens. And again, I've only dealt with digital projects less than 10 years old. Imagine what this will be like 30, 40, or 50 years from now.

If you think you're smarter than the Academy and have all the answers, I suggest you contact them now:

Academy of Motion Picture Arts and Sciences
Attn.: Science & Technology Council
1313 N. Vine Street
Hollywood, CA 90028
(310) 247-3000

I believe Mr. Andy Maltz is currently the director of the Science & Technology Council.

 

[Gavin Greenwalt]

The problem you overlook is that the LTO tapes and drives both can fail. And at the moment, this is all we have for longterm digital storage. I have seen multiple copies fail at different times, and it's not pretty when that happens. And again, I've only dealt with digital projects less than 10 years old. Imagine what this will be like 30, 40, or 50 years from now.

If your LTO drive fails... go buy another LTO drive, they aren't one of a kind of priceless objects, there is probably one on Craiglist somewhere nearby. I'm not overlooking this fact at all. I'm counting on both drives and tapes failing. But at some point where N = the number of copies you have, the chances of all your copies corrupting vs the chances of one film print being destroyed swings in favor of the multiple copies. I would argue that the value N is sufficiently low that yes multiple copies will fail, however multiple copies will also remain in tact. And finding a working copy among the failed copies will take less time than fixing all of the dust and registration problems with a film reel. Multiple tapes will die, but will 47 copies die in 10 years? You would have to be mad unlucky for something like that to happen. And if it were a common occurrence our banking industry would be in chaos.

In 30 years you'll have an easier time finding an LTO-6 deck than you will a working Arrilaser. The only people who will have film scanners in 30 years will be film archivists. The infrastructure is going to collapse for film scanning. Your local post house already probably sold or scrapped theirs. It's a large expensive niche device within a niche industry.

As to 30,40 or 50 years from now... our entire archives will fit on a thumb drive for $20.

30 years ago $1,500 bought you 5MB of data. Today $1,500 buys you 38,000,000 MB. The Digital Dilemma estimates that a 4k film needs about 10TB of data.

In 30 years for $1,500 you'll be able to buy 288,800 petabytes of data. To put that into perspective that's about 28,880,000 films in 4k. And it'll fit in a 2u rack server. If you archive your film today digitally its price of archival will fall logarithmically. Your price of film storage will steadily increase as the number of film scanners decrease and maintenance becomes harder and harder to perform on the existing (very rare) scanners.

 

[Ketch Rossi]

Not as a capture medium, but with regard to long term archival film is (and probably should remain) the only game in the business.

Unfortunately I also have to disagreed on this... Why?

I have visited several facilities which are dedicated to the long term Film archival, and all of them had the very same thing to say: We wish for something to replace this soon, very soon...

The reasons are far too many to list, going from some of them been hold format and high risks, which of course could be re-mastered with new stock, but too delicate to even touch, as for the new stoke, then there are all the long term issues starting right off with the continuos needs of refrigeration, were the facilities spend fortunes in AC to keep the stocks in a favorable temperature to allow for optimum long term storage.

But this are the miner ones, they are really some very big concerns which do not make Film as a good solution for any kind of use really, and long term storage the list of all.


Yet I must agree that currently of all solutions for any long term storage, Film remains one of the best, just looking forward to something much better, much safer, and much more secure to store important data for the along term.

 

[Elsie N]

I hear this over and over... it's even on Kodak's Advertisements--but again it's total nonsense. ZIP Disks are now 20 years old. It's proprietary and was never intended for long term archival. But it was a mass produced consumer product. As a result--it's trivial to find a working drive. Not only that you can literally buy them by the palette load. I could find a functional Zip Drive and hook it up by the end of the day. I wouldn't say that I could get a full scan of a separation archival film reel. When you produce millions of consumer devices there will almost always be cheap working examples. Need an 8 track player? No problem. It's 40 years old but there is probably one for sale within a mile radius of you right now that's in pristine condition for less than $100.

USB is now nearly 20 years old. And yet you can take any USB device from 1996 and plug it into almost every single computer in existence. FAT the file system has been around since the early 1980s. It's the formatting used on REDMags. Optical storage is an open ISO standard. I mean even the "O" in LTO tape is "Open". If you are truly worried than you can just archive a few sealed boxes in your vault.

Are they all unproven in the long term? Yes. Because they haven't existed in the long term. However, seat belts hadn't been proven in the long term either before we could conclude that they increased safety.

In my opinion it's the stone vs paper battle. The film industry is firmly entrenched in the "Stone" camp. They think it's worth carving it into stone since stone is strong. Stone is proven to last centuries, stone is expensive but good. I'm in the paper camp. Paper is flimsy. Paper burns. Paper is spoileable by water and animals. You would think stone is the way to go--but if you want Harry Potter to survive 1,000 years you don't put it on an elaborate stone tablet system, you print 10,000,000 copies and it'll just happen. Volume trumps strength.

I've also seen countless 20th Anniversary DVDs where they say the same thing every time. "The film was falling apart, if we didn't do it now we were going to lose it forever." I start from the assumption that people are going to screw up. And then based on that assumption what is most resistant to cock-ups? From the NASA archives to classic films the answer is always the exact same: some guy had a copy in his sock drawer or some TV station in Melbourne happened to have an old copy. Cockroaches are forever, not diamonds--diamonds get dropped onto the Titanic Wreckage never to be seen again.

Might be a good place to point out that R&D for new digital storage media is XXXX times the level for enhancing film for storage. So aim for what is to come rather than at this point on the timeline.

(Gavin, love your logic)

 

[dio Zafi]

here in Sundance 2014 camera survey from film playing here…

2 35mm (one arri and one panavision)
3 arri alexa
3 super 16mm
1 sony f 3
1 red epic

 

[M Most]

If your LTO drive fails... go buy another LTO drive...
In 30 years you'll have an easier time finding an LTO-6 deck than you will a working Arrilaser. ..
In 30 years for $1,500 you'll be able to buy 288,800 petabytes of data. To put that into perspective that's about 28,880,000 films in 4k. And it'll fit in a 2u rack server. If you archive your film today digitally its price of archival will fall logarithmically. Your price of film storage will steadily increase as the number of film scanners decrease and maintenance becomes harder and harder to perform on the existing (very rare) scanners.

Gavin, I'm not going to argue with you because you seem to be completely convinced that you understand it all and those who are dealing with it don't. So believe what you want.

 

[Elsie N]

One other point, whichever archival method is deemed the most likely to fail should continue to be used. There are a lot of films out there that really do need to deteriorate over time to the point of non-recovery. '-)

 

[David Kuhn]

I agree 100%

Digital archiving is still a pretty risky business despite the predictable degrading of analogue media.

I am sometimes playing head-games with what an archaeologist will do when he finds a CD, I know what they can deduct from a film-roll and an LP

That said, the largest volumes of art has always just faded.

Bach was not the Baroque period in music. He was a smalltime organist with a big family and a systematic mind. Mozart and Hayden were not the only "Classical" composers.

Time is a brutal editor of the past for us all.

On the same time, the majority of work (as in minutes made) is intended for consumption...

But it is an interesting thing.

There is a fair amount of effort going into the Digital Archive question... but as you mentioned with your archaeologist comment, it's a thorny issue. If you hand someone a CD in 100 years they certainly won't know what to do with it... but an expert will be able to quickly determine how the physical device stores the data. With effort, they will be able to extract the data and with more effort they'll be able to decode it. I think one of the things that all "Permanent" media meant for long term storage should do is include a header that uses a VERY simple format (ascii maybe) to store the data format of the device. With that you reduce the problem from a physical read + expensive encoding reverse engineer to a physical read + format decoder write (probably saving >50% of the project cost). The other solution of course is periodic maintenance... valuable data archives periodically updated to newest storage tech - expensive in time but not as much in hardware since whenever I've needed to upgrade I've been able to consolidate dramatically due to the decrease in cost/MB (20 years ago), cost/GB (10 years ago), cost/TB (2 years ago), cost/PB (couple years from now).

 

[Gavin Greenwalt]

Gavin, I'm not going to argue with you because you seem to be completely convinced that you understand it all and those who are dealing with it don't. So believe what you want.

I'm only completely convinced because nobody is offering an alternative perspective except "Well the Academy disagrees, who are you to disagree with the academy?" Clearly several of you disagree with me, so I would like to hear why I'm wrong not just more ad-hominem insinuation that I don't know what I'm talking about... because.

Let's just take one of my proposals such as LTO as an example. Let's say there is a 50% chance of failure over 10 years (very very poor tape quality or storage).
(1/2) ^ 4 copies = 12.5% chance of losing all 4 copies. Unacceptable high risk.
(1/2) ^ 10 copies = .097 % chance of failure with 10 copies. That means you would lose 1 in a thousand. Still too high.
(1/2) ^ 47 copies = This means you would lose 1 in about 10 trillion films.

This isn't a "new" problem the academy is facing. Data retention is a problem faced by nearly every industry. Every industry has managed to find a digital archival process that is affordable. And time and time again through history the answer has been the same: redundancy. Early jet liners had 4 engines because creating one super-reliable engine wasn't cost effective. Monks copied the Bible they didn't spend 1,000 years etching it into stone. The internet was designed to route around failure points.

 

[paulherrin]

gavin, you're spot on about stones & paper, gold & film (&) lost in 'the apocalypse'. i see immediately and easily-accessible archives on tape/drives/solid-state as better than a few films that need to be scanned and cleaned - this is in the short term. i do think and hope something more reliable will emerge for long-term digital storage in a single-instance. in either case, maintaining virtual copies (or some combo) seems like the best route imho - film is a physical technology, data is an information technology... which one has a better chance of being possible when it's far and removed from today and which one has a better chance of fading away from possibility?

i'm gonna try to read ab the dilemma, i'm not an expert in archival studies by any stretch and i realize there are different needs and different opinions, etc... i'd love to hear a discussion/debate on this fundamental logic/philosophy gg has so eloquently presented. maybe you should contact the academy experts

 

[Elsie N]

One other thing to consider when comparing film vs digital archival costs. I'm not talking about the Masters, but total footage. A truly great movie should be preserved in every way including Master, Director's cut, bloopers... and all the footage shot to allow a re-edit in the future when morays accept greater freedom of expression in the mainstream. This increases the amount of data or film to be saved by a large factor.

 

[Marc Wielage]

Gavin, I'm not going to argue with you because you seem to be completely convinced that you understand it all and those who are dealing with it don't. So believe what you want.

One other thing to consider when comparing film vs digital archival costs. I'm not talking about the Masters, but total footage. A truly great movie should be preserved in every way including Master, Director's cut, bloopers... and all the footage shot to allow a re-edit in the future when morays accept greater freedom of expression in the mainstream. This increases the amount of data or film to be saved by a large factor.

This is very true. A few years ago, I was involved with creating an extended "director's cut" of an existing $100M 2-hour feature, and it was a nightmare piecing together images from workprint, IPs, original camera negative, 3/4" tapes, and Avid files. In some scenes, the end result was a combination of all of the above. Trust me, 15-year-old data files do not look good very good in HD. Imagine an 8K world where we wind up having to uprez compressed HD (like something from Vimeo) and cutting it into a feature... not good.

While it's true that 10TB would probably cover an average feature, the problem really does boil down to all the different versions, rough cuts, alternate scenes, cut scenes, outtakes, bloopers, textless versions, plus subtitles, original soundtracks, original music scores, various different kinds of mixes... the sheer amount of files is stunning on many levels.

An old pal of mine in the late 1980s did some restoration work on Gone with the Wind, and we walked through MGM's warehouse and looked at the dozens and dozens of shelves they had just devoted to different 3-strip protection negatives, studio prints, interpositives, release prints, foreign elements, titles, soundtracks, outtakes, camera tests, and on and on and on. As he remarked, "there's over 100 shelves just devoted to this one title. Now, imagine making 25 movies a year, and your archives go back 80 years. And then there's TV shows on top of that. Now, try to store all that in one place." It's a very daunting thought when you're confronted with the reality of how vast one studio's archives can be, and then you consider there's dozens and dozens of studios, networks, and production companies throughout the world.